FL-Net
The federated network software behind Federated Learning Net: lets multiple data holders keep control over local data while participating in shared discovery, analysis, and governed workflows.
Collaboration for real-world scientific work
Federated Learning Net
Harmonized dataAudited workflowsPersistent networkIsolated, secure executionPrivacy-preserving federation
At a glance
One stack, every role
Data holdersDeploy your own client, no-code onboard data, and control access, participating in federated studies with ease.
Data ScientistsFind cohorts, run (federated) analyses, and review and publish governed results.
Tool developersPackage scientific solutions as reusable Tools.
Collaboration coordinatorsSelf-host a FL-Net deployment
Know the pieces
The documentation is easier to navigate when the terminology is explicit. These are the building blocks used throughout the site.
Federated Learning Net
The deployed instance of FL-Net: a ready-to-use platform for finding data, running tools, and collaborating without building infrastructure from scratch.
Client Deployments
Self-hosted nodes operated by data holders which make up the Federated Learning Net. They enforce access rules, run no-code data onboarding, and participate in federated analysis.
Jump into the documentation
Orient yourself with the pages above, or go straight to the path that matches your role.
Welcome
Understand what FL-Net and Federated Learning Net are and how the pieces fit together.
Open sectionSolutions
See why teams choose FL-Net over other federated learning frameworks.
Open sectionGet Started
Choose the right path based on your role and deployment model.
Open sectionArchitecture
Understand how the platform, clients, and runtime boundaries fit together.
Open sectionDeploy your own client
Set up a self-hosted client, either on your own infrastructure or in a cloud environment.
Open sectionUse your client
Describe your schema, ingest data through connectors, and decide what the network is allowed to do with it.
Open sectionUse the platform
Search the network, create analysis projects, run tools, and interpret results in a governed environment.
Open sectionBuild and publish tools
Develop reproducible ETL, federated learning, and inference Tools for others to run.
Open sectionSelf-host a deployment
Run your own platform and coordinate the clients that connect to it.
Open sectionContribute to FL-Net
Help build the underlying software: the Client, Platform, and Schema system itself.
Open sectionTypical workflow
FL-Net is designed so discovery, governance, execution, and interpretation follow a consistent sequence instead of turning into manual coordination.
01
Describe and publish metadata
Data holders model their local schema and expose only the information needed for discovery and governance.
02
Discover relevant data and tools
Data Scientists search the network, evaluate available tools, and assemble a project that matches their question.
03
Run governed analysis
Execution happens through reproducible tools and explicit permissions rather than ad hoc data movement.
04
Review outputs and iterate
Results, logs, and artifacts stay traceable, so teams can refine parameters and compare outcomes.